Project Summary Rapid Electrode Multiplexing for Scalable Neural Recording Large-scale recording of neural signals is essential for gaining a better understanding of the elaborate, dynamic picture of the brain that emerges from interactions involving individual cells and complex neural circuits. Over the past few decades extracellular neural recording capabilities have progressed from single unit in vitro recordings to simultaneous monitoring of the activity of about one hundred neurons in vivo. Currently, microwire and microfabricated silicon neural probes are capable of sensing the activity of 100's of neurons. Miniaturized recording systems based on CMOS integrated circuits (ASICs) have been developed that can record from 16- 100 neurons simultaneously, and these chips are no bigger than a postage stamp. The miniaturization of these recording circuits, however, has not kept pace with advances in neural probe technology. To access 1000+ channels of neural activity, surgical implications require aggressively miniaturized fully implantable recording ASICs, and signals must be multiplexed to reduce percutaneous wire counts. Given the miniscule dimensions constraining implantable devices, existing IC designs cannot be simply `copy/pasted' to scale to orders of magnitude higher channel counts. The neuroscience community greatly needs a rapidly multiplexed circuit architecture in which the analog interfaces are shared by many recording sites, enabling orders of magnitude smaller area per channel. This project will collect in vivo microelectrode characterization data to support a new class of rapidly multiplexed ASICs, and will demonstrate the new ASIC paradigm with a prototype chip.